Development of a control system for gas concentration and temperature in an animal preference chamber

Environmental Control;Animal Preference;Gas Concentration;Animal Welfare;Fuzzy Logic

Johnson, Ryan ; Green ; Angela R.

关键词: Environmental Control;    Animal Preference;    Gas Concentration;    Animal Welfare;    Fuzzy Logic;   

Others  :  https://www.ideals.illinois.edu/bitstream/handle/2142/45519/Ryan_Johnson.pdf?sequence=1&isAllowed=y

美国|英语

来源: The Illinois Digital Environment for Access to Learning and Scholarship

【 摘 要 】

An enhancement of the environmental control capabilities was planned and executed for an Environmental Preference Chamber (EPC) used for animal behavior testing. A fuzzy logic controller (FLC) was designed and built to control distinct ammonia concentration levels independently in each of the EPC’s four compartments. The FLC was written in MATLAB and utilized in tandem with custom software created in a visual-based programming language. The control software compared NH3 measurements from an infrared photo-acoustic gas analyzer to user-defined setpoints for each compartment, and input the difference to the FLC. The FLC computedan incremental change to the voltage signal used to adjust four Mass Flow Controllers, changing the volumetric flow rate of supplied NH3 to each compartment. Average (± standard deviation) NH3 concentrations were: 1.8 ± 0.8ppm, 10.2 ± 0.5ppm, 20.1 ± 0.8ppm, and 40.5 ± 1.3ppm for setpoint concentrations of 0, 10, 20, and 40ppm, in each of the four compartments, respectively. Approximately 90 minutes were required for all compartments to be within 5% of their setpoint concentration when starting from fresh air conditions.An expansion of the heating capacity of each compartment of the Environmental Preference Chamber was performed by increasing the installed heat capacity of in-line electrical resistance coils from 200W to 600W. The maximum temperature rise increased from 5.7±0.5°C to 15.1±0.6°C, with a time constant, τ, of 1.3±0.1 h. The 600W heating capacity achieved 95% of a 1°C and 3°C positive temperature step response in 12.3±2.3 min and 24.0±3.3 min, respectively. This was faster than the original 200W heating system, which achieved 95% of a 1°C and 3°C positive temperature step response in 18.0±2.8 min and 51.9±12.8 min, respectively. The setpoint overshoot of the new 600W heating system was 1.5±0.4°C and 0.9±0.1°C for 1°C and 3°C, respectively. This overshoot was greater than for the original 200W heating system, which showed a 0.4±0.1°C overshoot for the +1°C step response and a negligibly small overshoot for the 3°C. Cooling to room temperature was achieved using supply fans. First-order step response time constants, τ, for 1°C and 3°C negative step responses to ambient air temperature were 39.0±3.4 min and 26±1.4 min, respectively. The discrete On/Off temperature controller did not noticeably impact the performance of the fuzzy logic gas concentration controller when operating simultaneously. Four distinct ammoniated environments were created while the temperature controller was set to maximum temperature rise. The ammonia concentrations experienced minor perturbations during temperature rise, but the gas concentration controller quickly corrected these. The exhaust fan created noise in the Temp/Rh sensors, affecting the temperature control by dampening any oscillations when achieving a setpoint of 27°C(temperature increase of +5°C above ambient air temperature). No meaningful difference was observed in the gas concentration controller performance when simultaneously maintaining a temperature setpoint.